MemoryManager¶

MemoryManager is an abstraction of memory managers that can share available memory between task execution (TaskMemoryManager) and storage (BlockManager).

MemoryManager splits assigned memory into two regions:

• Execution Memory for shuffles, joins, sorts and aggregations

• Storage Memory for caching and propagating internal data across Spark nodes (in on- and off-heap modes)

MemoryManager is used to create BlockManager (and MemoryStore) and TaskMemoryManager.

Contract¶

Acquiring Execution Memory for Task¶

acquireExecutionMemory(
  numBytes: Long,
  taskAttemptId: Long,
  memoryMode: MemoryMode): Long

Used when:

• TaskMemoryManager is requested to acquire execution memory

Acquiring Storage Memory for Block¶

acquireStorageMemory(
  blockId: BlockId,
  numBytes: Long,
  memoryMode: MemoryMode): Boolean

Used when:

• MemoryStore is requested for the putBytes and putIterator

Acquiring Unroll Memory for Block¶

acquireUnrollMemory(
  blockId: BlockId,
  numBytes: Long,
  memoryMode: MemoryMode): Boolean

Used when:

• MemoryStore is requested for the reserveUnrollMemoryForThisTask

Total Available Off-Heap Storage Memory¶

maxOffHeapStorageMemory: Long

May vary over time

Used when:

• BlockManager is created
• MemoryStore is requested for the maxMemory

Total Available On-Heap Storage Memory¶

maxOnHeapStorageMemory: Long

May vary over time

Used when:

• BlockManager is created
• MemoryStore is requested for the maxMemory

Implementations¶

• UnifiedMemoryManager

Creating Instance¶

MemoryManager takes the following to be created:

• SparkConf
• Number of CPU Cores
• Size of the On-Heap Storage Memory
• Size of the On-Heap Execution Memory

Accessing MemoryManager (SparkEnv)¶

MemoryManager is available as SparkEnv.memoryManager on the driver and executors.

import org.apache.spark.SparkEnv
val mm = SparkEnv.get.memoryManager

scala> :type mm
org.apache.spark.memory.MemoryManager

Associating MemoryStore with Storage Memory Pools¶

setMemoryStore(
  store: MemoryStore): Unit

setMemoryStore requests the on-heap and off-heap storage memory pools to use the given MemoryStore.

setMemoryStore is used when:

• BlockManager is created

Execution Memory Pools¶

On-Heap¶

onHeapExecutionMemoryPool: ExecutionMemoryPool

MemoryManager creates an ExecutionMemoryPool for ON_HEAP memory mode when created and immediately requests it to incrementPoolSize to onHeapExecutionMemory.

Off-Heap¶

offHeapExecutionMemoryPool: ExecutionMemoryPool

MemoryManager creates an ExecutionMemoryPool for OFF_HEAP memory mode when created and immediately requests it to incrementPoolSize to...FIXME

Storage Memory Pools¶

On-Heap¶

onHeapStorageMemoryPool: StorageMemoryPool

MemoryManager creates a StorageMemoryPool for ON_HEAP memory mode when created and immediately requests it to incrementPoolSize to onHeapExecutionMemory.

onHeapStorageMemoryPool is requested to setMemoryStore when MemoryManager is requested to setMemoryStore.

onHeapStorageMemoryPool is requested to release memory when MemoryManager is requested to release on-heap storage memory.

onHeapStorageMemoryPool is requested to release all memory when MemoryManager is requested to release all storage memory.

onHeapStorageMemoryPool is used when:

• MemoryManager is requested for the storageMemoryUsed and onHeapStorageMemoryUsed
• UnifiedMemoryManager is requested to acquire on-heap execution and storage memory

Off-Heap¶

offHeapStorageMemoryPool: StorageMemoryPool

MemoryManager creates a StorageMemoryPool for OFF_HEAP memory mode when created and immediately requested it to incrementPoolSize to offHeapStorageMemory.

MemoryManager requests the MemoryPools to use a given MemoryStore when requested to setMemoryStore.

MemoryManager requests the MemoryPools to release memory when requested to releaseStorageMemory.

MemoryManager requests the MemoryPools to release all memory when requested to release all storage memory.

MemoryManager requests the MemoryPools for the memoryUsed when requested for storageMemoryUsed.

offHeapStorageMemoryPool is used when:

• MemoryManager is requested for the offHeapStorageMemoryUsed
• UnifiedMemoryManager is requested to acquire off-heap execution and storage memory

Total Storage Memory Used¶

storageMemoryUsed: Long

storageMemoryUsed is the sum of the memory used of the on-heap and off-heap storage memory pools.

storageMemoryUsed is used when:

• TaskMemoryManager is requested to showMemoryUsage
• MemoryStore is requested to memoryUsed

MemoryMode¶

tungstenMemoryMode: MemoryMode

tungstenMemoryMode tracks whether Tungsten memory will be allocated on the JVM heap or off-heap (using sun.misc.Unsafe).

tungstenMemoryMode is OFF_HEAP when the following are all met:

• spark.memory.offHeap.enabled configuration property is enabled

• spark.memory.offHeap.size configuration property is greater than 0

• JVM supports unaligned memory access (aka unaligned Unsafe, i.e. sun.misc.Unsafe package is available and the underlying system has unaligned-access capability)

Otherwise, tungstenMemoryMode is ON_HEAP.

tungstenMemoryMode is used when:

• MemoryManager is created (and initializes the pageSizeBytes and tungstenMemoryAllocator internal properties)
• TaskMemoryManager is created

MemoryAllocator¶

tungstenMemoryAllocator: MemoryAllocator

MemoryManager selects the MemoryAllocator to use based on the MemoryMode.

tungstenMemoryAllocator is used when:

• TaskMemoryManager is requested to allocate a memory page, release a memory page and clean up all the allocated memory